TinyOL: TinyML with Online-Learning on Microcontrollers
| Autor: | Thomas A. Runkler, Darko Anicic, Haoyu Ren |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Flexibility (engineering)
FOS: Computer and information sciences Computer Science - Machine Learning Artificial neural network business.industry Computer science Computation Deep learning Inference Systems and Control (eess.SY) Machine learning computer.software_genre Electrical Engineering and Systems Science - Systems and Control Autoencoder Machine Learning (cs.LG) Microcontroller Memory management Computer Science - Distributed Parallel and Cluster Computing FOS: Electrical engineering electronic engineering information engineering Artificial intelligence Distributed Parallel and Cluster Computing (cs.DC) business computer |
| Zdroj: | IJCNN |
| DOI: | 10.48550/arxiv.2103.08295 |
| Popis: | Tiny machine learning (TinyML) is a fast-growing research area committed to democratizing deep learning for all-pervasive microcontrollers (MCUs). Challenged by the constraints on power, memory, and computation, TinyML has achieved significant advancement in the last few years. However, the current TinyML solutions are based on batch/offline settings and support only the neural network's inference on MCUs. The neural network is first trained using a large amount of pre-collected data on a powerful machine and then flashed to MCUs. This results in a static model, hard to adapt to new data, and impossible to adjust for different scenarios, which impedes the flexibility of the Internet of Things (IoT). To address these problems, we propose a novel system called TinyOL (TinyML with Online-Learning), which enables incremental on-device training on streaming data. TinyOL is based on the concept of online learning and is suitable for constrained IoT devices. We experiment TinyOL under supervised and unsupervised setups using an autoencoder neural network. Finally, we report the performance of the proposed solution and show its effectiveness and feasibility. Comment: Accepted by The International Joint Conference on Neural Network (IJCNN) 2021 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|